[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1831224a48dfbf54fb45fa56fce826d1d312700f.camel@redhat.com>
Date: Mon, 06 Nov 2023 08:21:52 +0100
From: Paolo Abeni <pabeni@...hat.com>
To: Eric Dumazet <edumazet@...gle.com>, Mat Martineau <martineau@...nel.org>
Cc: Matthieu Baerts <matttbe@...nel.org>, "David S. Miller"
<davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>,
netdev@...r.kernel.org, mptcp@...ts.linux.dev
Subject: Re: [PATCH net-next 9/9] mptcp: refactor sndbuf auto-tuning
Hi,
On Thu, 2023-11-02 at 18:19 +0100, Eric Dumazet wrote:
> On Mon, Oct 23, 2023 at 10:45 PM Mat Martineau <martineau@...nel.org> wrote:
> >
> > From: Paolo Abeni <pabeni@...hat.com>
> >
> > The MPTCP protocol account for the data enqueued on all the subflows
> > to the main socket send buffer, while the send buffer auto-tuning
> > algorithm set the main socket send buffer size as the max size among
> > the subflows.
> >
> > That causes bad performances when at least one subflow is sndbuf
> > limited, e.g. due to very high latency, as the MPTCP scheduler can't
> > even fill such buffer.
> >
> > Change the send-buffer auto-tuning algorithm to compute the main socket
> > send buffer size as the sum of all the subflows buffer size.
> >
> > Reviewed-by: Mat Martineau <martineau@...nel.org>
> > Signed-off-by: Paolo Abeni <pabeni@...hat.com>
> > Signed-off-by: Mat Martineau <martineau@...nel.org
>
> ...
>
> > diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
> > index df208666fd19..2b43577f952e 100644
> > --- a/net/mptcp/subflow.c
> > +++ b/net/mptcp/subflow.c
> > @@ -421,6 +421,7 @@ static bool subflow_use_different_dport(struct mptcp_sock *msk, const struct soc
> >
> > void __mptcp_set_connected(struct sock *sk)
> > {
> > + __mptcp_propagate_sndbuf(sk, mptcp_sk(sk)->first);
>
> ->first can be NULL here, according to syzbot.
I'm sorry for the latency on my side, I had a different kind of crash
to handle here.
Do you have a syzkaller report available? Or the call trace landing
here?
Thanks!
Paolo
Powered by blists - more mailing lists